Download Forecasting pollination declines through DNA barcoding

Forum
Viewpoints
Forecasting pollination declines
through DNA barcoding: the
potential contributions of
macroecological and
macroevolutionary scales of
inquiry
Summary
While pollinators are widely acknowledged as important contributors to seed production in plant communities, we do not yet have a
good understanding of the importance of pollinator specialists for
this ecosystem service. Determination of the prevalence of pollinator specialists is often hindered by the occurrence of cryptic species
and the limitations of observational data on pollinator visitation
rates, two areas where DNA barcoding of pollinators and pollen can
be useful. Further, the demonstrated adequacy of pollen DNA
barcoding from historical records offers opportunities to observe the
effects of pollinator loss over longer timescales, and phylogenetic
approaches can elucidate the historical rates of extinction of
specialist lineages. In this Viewpoint article, we review how
advances in DNA barcoding and metabarcoding of plants and
pollinators have brought important developments to our understanding of specialization in plant–pollinator interactions. We then
put forth several lines of inquiry that we feel are especially promising
for providing insight on changes in plant–pollinator interactions over
space and time. Obtaining estimates of the effects of reductions in
specialists will contribute to forecasting the loss of ecosystem
services that will accompany the erosion of plant and pollinator
diversity.
Introduction
Specialization in plant–pollinator interactions is a field of study
that, like many in biology, is plagued with hidden players and
cryptic mechanisms (e.g. pollen is small, and the act of pollen
delivery is difficult to evaluate with the human eye; Vamosi et al.,
2012). Further, on large geographical scales, the speed and cost of
the necessary detailed measurements are often prohibitive. Studies
of plant–pollinator interactions require the accurate identification
of both pollinators and the pollinated, and recent efforts with DNA
Ó 2016 The Authors
New Phytologist Ó 2016 New Phytologist Trust
barcoding has demonstrated success in achieving this objective
without prolonged observations. Landscape-level studies of the
effects of changes in pollinator composition on ecosystem health
are exceedingly rare (Fr€
und et al., 2013; Tur et al., 2013) because of
the difficulties in getting sample sizes needed to make firm
conclusions. Plant–pollinator network studies often rely on
comparing a small number of communities (e.g. two), and even
these require >100 h of visitor observation (Tur et al., 2013).
Estimating the effects of land use on the disappearance of specialists
can require a herculean effort, such as the recent study that
conducted 962 h of surveying of pollinator visitation in 119
grassland sites with varying levels of disturbance (Weiner et al.,
2014), yet such studies will become increasingly necessary if we
wish to evaluate the effect of species loss on ecosystem services.
Here, we address how DNA barcoding can vastly reduce the effort
of these macroecological studies.
The field hours documenting plant–pollinator interactions do
not account for the hours spent in the laboratory identifying
pollinators, and DNA barcoding is well established to assist with
this endeavor (Sheffield et al., 2009). For example, bees and other
insects can be identified with great accuracy using the standard
animal barcode CO1 fragment (Hebert et al., 2003b), and a survey
of European bees found that results from DNA barcoding largely
agreed with traditional taxonomy (Schmidt et al., 2015). Further,
other studies reveal that morphologically indistinguishable (i.e.
cryptic) species can be differentiated through DNA barcode
markers (Smith et al., 2006; Schmidt et al., 2015) or other
molecular methods (e.g. the cryptic species group previously
known solely as Halictus ligatus; Packer et al., 2016) although the
converse is also occasionally true (Gibbs, 2010).
Identifying insects through DNA barcoding has become
relatively standard in the field, yet DNA barcoding the pollen
found on pollinators is a more recent development. The situation is
not as simple with plants because at least two gene fragments,
typically rbcL and matK, are required to obtain levels of accuracy
above 70% (Pei et al., 2015). Thus, species-level discrimination is
more difficult for closely-related species of plants, especially in
instances where hybridization occurs (Kress et al., 2005; Clement
& Donoghue, 2012). While the use of plastid markers was
originally assumed to be inadequate for pollen (thought to be
without chloroplasts), several studies have found this not to be the
case, expanding the toolbox for pollen DNA barcoding further
(reviewed in Bell et al., 2016). With both plants and pollinators,
extensive barcode libraries (obtained from herbaria and museum
collections) are required to permit ecological samples to be
identified. Methodological advances are needed to obtain full
species-level resolution when DNA barcoding just the pollen, such
as sequencing whole chloroplast genomes or incorporating novel
blended approaches between few-marker barcoding and organellar
genomics (e.g. Li et al., 2015; Coissac et al., 2016; Hollingsworth
New Phytologist (2016) 1
www.newphytologist.com
2 Forum
Viewpoints
et al., 2016). Nevertheless, recent surveys highlight the rapid
development of DNA barcode libraries, with the vast majority of
barcodes in BOLD (Ratnasingham & Hebert, 2007; CBOL Plant
Working Group, 2009) being for animals (4.30M/406K/154K
specimens/BINS/species with publicly available CO1 sequences),
followed by plants (301K public specimen records representing
84K species, mainly with rbcL sequences but other markers as well;
BOLD accessed 26 October, 2016). Here, we explore how these
new techniques can expand the scope of previous macroecological
investigations on pollination to aid investigations of how specialization evolves and how specialists contribute to ecosystem
function.
Metabarcoding is a novel technique that allows multispecies
assemblages to be identified from a single sample without
separating out individual pollen grains (Box 1). The expanding
DNA libraries have been demonstrated capable of reconstructing
the plants visited by pollinators simply through DNA barcoding of
the pollinator and the pollen on the pollinator’s body (Widmer
et al., 2000), despite the mixed sample of plant pollen commonly
retrieved from polylectic pollinators (reviewed in Bell et al., 2016).
Metabarcoding has been used to characterize the composition and
relative abundance of pollen collected by honeybees based upon the
species of pollen found in scopal loads (Richardson et al., 2015a).
These methods have already been used to determine how
pollination of a single pollinating species changes with local
floristic biodiversity, plant phenology, and the presence of alien
flowering species (Wilson et al., 2010; Galimberti et al., 2014),
elucidating the potential for macroecological comparisons of
pollination over large spatial scales with substantial time savings. A
third study identified 19 plant families from honeybee scopal
pollen loads and showed that metabarcoding exhibited greater
taxonomic sensitivity in large and diverse pollen samples relative to
microscopy, which found only eight families (Richardson et al.,
2015b), a result that has stood up to scrutiny in other systems that
have taken pains to control for the possibility of false positives
through trace contaminants (Pornon et al., 2016). Additionally, it
has been suggested that metabarcoding is substantially less
expensive than traditional barcoding or morphological identification when person hours are included in the costs (Tang et al., 2015).
This is especially the case in geographic areas where taxonomic
expertise is sparse or for taxonomic groups in which pollen grains
are extremely difficult to identify morphologically even to genus
level (Schuett & Vamosi, 2010). Of course, in both cases, a locally
extensively barcoded fauna/flora is essential to make species-level
designations possible and represents a substantial investment.
We argue that investment in these barcode libraries as well as
further development of methods would provide worthwhile gains
for understanding plant–pollinator specialization in ecosystems
and outline two areas where recent developments provide windows
into future opportunities. First, the costs associated with DNA
barcoding and morphological identification are still outside the
budgets of most researchers, yet we feel the currently decreasing
costs of DNA barcoding will potentially tip the scales towards the
methodology should the increased efficiencies continue (Box 1).
One recent study was successful at multiplexing 384 pollen samples
collected from solitary bees and sequenced all samples together on a
New Phytologist (2016)
www.newphytologist.com
New
Phytologist
single Illumina MiSeq v2 flow cell, revealing 650 different plant
taxa visited (of which 617 could be identified taxonomically to
plant species level). Because samples were tagged before multiplexing, the suite of plants visited could be determined for each
individual sample (Sickel et al., 2015). For comparison, a recent
field study that included a similar amount of effort in terms of
collecting the samples of pollinators had to limit their surveys to
estimate the visitors to only 61 identified plant species (an order of
magnitude fewer) through traditional identification of pollinators
(Popic et al., 2013). Pantrap samples can also be analyzed with
metabarcoding to discover both the suite of pollinators obtained
(Tang et al., 2015) as well as the aggregate of pollen they carried
with them into the trap.
Because pollinators can visit flowers without contacting the
stigmas of plants, what remains to be developed are methods of
sampling the composition of pollinators that visited an individual
flower directly from trace amounts of pollinator DNA left behind
in nectar or on flower surfaces. This type of metabarcoding can be
thought of as an environmental DNA (eDNA) approach, which is a
promising new method where the identities of a multispecies pool
of plant species is recovered from trace amounts of DNA in an
environmental sample (Richardson et al., 2015b). For example,
there has been success in identifying eDNA of bees from samples of
honey (Schnell et al., 2010). Further, some success has been
obtained with reconstructing the suite of pollinators of focal plant
species from DNA barcoding of ‘microbial signatures’ in nectar and
flower surfaces with that found on particular pollinators. While this
practice is still in its infancy and will require further testing
(Aizenberg-Gershtein et al., 2013; Ushio et al., 2015), it would
circumvent the need for time spent capturing pollinators. Later, we
explore how DNA barcoding holds promise to uncover the
ecosystem function and conservation importance of pollinator
diversity at macroecological scales of inquiry.
DNA barcoding and the potential for understanding
plant–pollinator interactions
If we consider the success in developing methods of pollen DNA
metabarcoding from sampled pollinators (Bell et al., 2016), these
data can be used to construct more rapid estimates of pollinator
specialization within communities spanning a large geographical
gradient (Box 1). Not only is there the potential for broader
sampling, there is the potential for greater accuracy in our estimates
of specialization, and these changes should improve our biological
inferences regarding mutualisms. For example, DNA barcode
markers have demonstrated that many apparent dietary generalist
insect species are actually large numbers of specialist cryptic species
(e.g. Smith et al., 2011). Similarly, community-level metrics of
specialization can change dramatically with DNA barcode data
(Clare, 2014; Toju et al., 2014; Roslin & Majaneva, 2016).
Generally, there are two sources of errors with traditional
identifications as a basis of concluding species interactions: (1)
actual misidentifications, especially in large studies with a lot of
specimens to identify and (2) cases of missing observations, e.g. due
to rare or cryptic species being overlooked. In some systems (e.g.
tropics), such studies would not be possible at all without molecular
Ó 2016 The Authors
New Phytologist Ó 2016 New Phytologist Trust
New
Phytologist
methods, as the majority of insect species remain undescribed and
are certainly unidentifiable in the field (Stork et al., 2015). With
DNA barcoding providing valuable information, such as the pollen
on a given pollinator species (or Molecular Operational Taxonomic
Unit) (Galimberti et al., 2014), we are afforded a more accurate
estimation of pollinator diet breadth when cryptic species of
pollinators are identified (Box 1). Should wide discrepancies be
seen between the visually observed visitation patterns and the pollen
present on visitors, these methods offer a way of identifying efficient
pollinators (Table 1) (Popic et al., 2013).
Identifying pollinator effectiveness, however, will require still
more effort in the form of examining whether visitation rates,
pollinator identity, and pollen composition on a pollinator’s body
correspond with conspecific pollen deposition to, and subsequent
seed production of, a focal species (Kremen et al., 2002). While the
scope of many macroecological questions regarding plant–pollinator interactions can be addressed with DNA barcoding (Box 1), it
is important to delineate where and when DNA barcoding
approaches will be useful and where they will not. For example,
DNA barcoding cannot separate self from nonself pollen: different
markers are required for discrimination among individuals of a
species (e.g. microsatellites or AFLP). For much of pollination
biology involved with the study of inbreeding depression, selfing
rates, and the concomitant effects on the evolution of dioecy and
floral traits (e.g. inflorescence structure) (Lloyd, 1982; Harder &
Barrett, 1995), DNA barcoding techniques will be of limited
utility. Similarly, understanding specialization of any given species
at a population level will also still benefit from traditional ecological
approaches because the ability to discern pollen abundance vs
presence/absence is still in its infancy (yet there is promise there as
well; Tang et al., 2015; Pornon et al., 2016). However, for the
effects of pollinator species richness and composition on the
sustainability of plant species and communities over macroecological scales, DNA barcoding offers a wealth of opportunities
(Table 1). We specifically highlight later several approaches to
incorporating barcode information into the study of the ecosystem
services provided by specialists in plant–pollinator systems.
Variation in specialization and pollination services
Over macroecological and macroevolutonary scales, the designation of specialization as a species trait is complicated because
specialization can change over both space and time, and there are
logistical barriers to conducting numerous surveys of plant–
pollinator interactions over a large number of sites or years.
Nevertheless, even our curtailed sampling to date demonstrates that
the extent of specialization observed can depend heavily on how
often alternate partners are encountered, which depends on the
relative abundance and range overlap with potential mutualists
(Sjodin, 2007; Vamosi et al., 2014a). While high pollinator species
diversity is often argued to offer benefits in the form of pollination
services to plant species (Kremen et al., 2002), the mechanisms
underlying this important biodiversity-ecosystem service (BES)
relationship are still poorly understood. Due to the ephemeral
nature of specialist lineages over space and time, the presence of
specialists is often assumed to be of limited importance in the
Ó 2016 The Authors
New Phytologist Ó 2016 New Phytologist Trust
Viewpoints
Forum 3
delivery of pollination services (Memmott et al., 2004) and may,
therefore, not greatly impact plant reproduction and extinction.
However, recent studies that have examined pollen delivery instead
of visitation have found that the disappearance of specialists
impacts the foraging behavior of the remaining generalists and, in
turn, the pollen delivery to the plant community (Brosi & Briggs,
2013). DNA barcoding techniques offer expanded opportunities
for the large-scale study of plant–pollinator specialization such that
we can: (1) determine how pollinator diversity relates to pollendelivery adequacy of a plant community; and (2) refine measurements of specialization for both the pollinators and the plants and
examine how these change across gradients (Table 1). We describe
these two avenues of inquiry below.
Pollinator specialization runs a broad spectrum from extreme
specialization of a plant species on a single pollinator species to
incorporating upwards of > 100 mutualist species (Vamosi et al.,
2013), yet evidence for whether higher species diversity of
pollinators is beneficial depends on the focal plant species in
question (e.g. Kremen et al., 2002; Gomez et al., 2007; Davila
et al., 2012). Meta-analysis indicates that specialization is generally
risky; plants visited by many pollinators (> 5 species) are less pollen
limited than those visited by few (1–5) species (Knight et al., 2005).
The effects of pollinator diversity on plant reproduction may be
obscured because visitation rates are often not an adequate
reflection of pollination rates because pollinators vary in their
effectiveness (e.g. King et al., 2013), which could underlie why
visitation rates do not often correspond with the level of pollen
limitation (Muir & Vamosi, 2015). Previous attempts to tease
actual pollination apart from observations of visits have revealed
large differences in pollen transport networks vs visitation networks
(Ballantyne et al., 2015), yet these studies have had to resort to
painstaking visual pollen identification procedures that require rare
alpha taxonomic expertise. Although there are automated visual
approaches, they are also expensive, time consuming, and need just
as extensive a reference database as does DNA barcoding (Marcos
et al., 2015). Abundance and composition of heterospecific pollen
transfer (HPT) has been determined to affect seed production and
may underlie why oligolectic (specialist) pollinators heighten plant
fitness more so than generalists (Arceo-Gomez & Ashman, 2011;
Brosi & Briggs, 2013). If stigmas were collected after single visits
from a variety of pollinators, it would be possible to gain estimates
of each pollinator’s mean quality of pollen deposition (in terms of
the frequency of pure conspecific pollen loads identified through
DNA barcoding).
While specialists likely contribute some level of pollination
services in any given community, the labile nature of specialization
complicates our interpretation of what elements of diversity we
should strive to conserve. The ecosystem service of pollination is a
community-level process, and therefore maintaining it may be
difficult to reconcile with the current system of prioritizing species
for conservation. The designation of species-at-risk is based on
species attributes (e.g. range and population size) and may therefore
fail to capture the functional role of a species (i.e. such habitat
specialists are often diet generalists (Litsios et al., 2014)). By
necessity, specialization is often considered a relatively invariant
species-property that is consistent across locations (Devictor et al.,
New Phytologist (2016)
www.newphytologist.com
4 Forum
New
Phytologist
Viewpoints
Box 1 Examples of how barcodes can be used to examine plant–pollinator interactions.
Pictures show (a) pollen of Dipsacus asperoides on the body of Bombus festivus, and (b) nectar extraction from Delphinium tenii. Plant species
visited by a single pollinator can be determined with DNA metabarcoding, but reconstructing the list of pollinator species that visited a given
plant from DNA barcode fragments in nectar is still in development. Obtaining these measurements for all floral visitors collected at a site,
individual and community-level metrics of specialization can be constructed. Pairing these metrics with seed production values offers ways to
determine what pollinator species, or combinations of species, play the greatest functional roles in communities. DNA barcoding permits these
metrics to be evaluated at unprecedented spatial and temporal scales, especially if barcodes (even that of only cytochrome oxidase 1 (CO1)) is
added to that of backbone phylogenies. Using historical ecology approaches will allow us to estimate whether the most critical pollinators in any
given system are declining.
(a)
(b)
2010). Models that have incorporated spatial heterogeneity find
that specialists should evolve when their habitat or mutualists
become common (Whitlock, 1996; Debarre & Gandon, 2010).
New Phytologist (2016)
www.newphytologist.com
There have been few empirical tests of these models using hostplant use because they require community-level sampling over vast
areas (Bridle et al., 2013; Vamosi et al., 2014a; Adderley & Vamosi,
Ó 2016 The Authors
New Phytologist Ó 2016 New Phytologist Trust
New
Phytologist
Viewpoints
Forum 5
Table 1 Summary of questions where DNA barcoding should prove useful in broadening the scope and precision of studies focused on understanding the
relationship between diversity and pollination ecosystem services
Specialization question
Advantages of approach with DNA barcoding
Demonstrated potential
How does the level of specialization change when
measured through pollen on pollinators vs visually
observed visitors?
Examine whether pollen of plants visited is actually
transported on insects by DNA barcoding pollen
found on insects.
Advantage: identifying pollen through
microscopy is a labor-intensive process requiring
extensive expertise; identifying to species-level is
often not possible and has accuracy < 78%
(Mander et al., 2014) as opposed to ≤ 95% with
pollen DNA barcoding
Standard DNA barcoding of plant community with
multiplexed DNA barcoding of pollen sampled
from pollinators and/or stigmas.
Advantage: thus far, macroecological studies
that follow species often reduce sampling to a
very small number of communities, species, or
individuals (e.g. Tur et al., 2013; with 27 species,
c. three individuals each, in two communities or
162 sampling units, whereas a single multiplexed
run of pollen samples can contain double this
number)
Barcoding pollen on plant and pollinator specimens
in collections and determine which interactions
have been lost through time.
Advantage: as mentioned earlier, DNA pollen
barcoding removes the prohibitive time barrier of
identifying the pollen through microscopy
Sickel et al. (2015); Richardson et al. (2015a);
Galimberti et al. (2014)
How does pollinator sharing (i.e. specialization)
change within a species’ range (or as a function of
pollinator species richness or disturbance)?
Is the rate of turnover of specialists higher than
generalists in disturbed areas; is it elevated
compared to historical levels?
2015). By assessing the pollen on pollinators through DNA
barcoding across a species’ range, we can determine to what degree
this heterogeneity in ecological options throughout the geographic
range of a species determines our measurement of pollinator
specialization (Table 1). By accurately measuring the level of
context-dependent ecological specialization in plant–pollinator
interactions over longer timescales, we can examine how constancy
over time has affected the evolution of pollinator specialization and
determine the effects that losing specialists will have on pollination
services.
Temporal variation in specialization and the stability
of ecosystem services
While some heterogeneity in specialization occurs naturally,
intensive human impacts result in a decrease in the degree of
specialization in communities over time (Weiner et al., 2014).
Habitat fragmentation has been repeatedly observed to reduce
pollination services and plant reproductive success (Knight et al.,
2005). Evidence suggests that specialists often decline with
disturbance more readily than generalists (Packer et al., 2005;
Zayed et al., 2005; Aguilar et al., 2006; Weiner et al., 2014). High
nestedness of networks (the degree to which specialists employ
subsets of the suite of species used by generalists) has been used as an
indicator of greater robustness of networks (Burkle et al., 2013).
Empirical observations indicate that nestedness increases when
specialists are present in a community, suggesting the loss of
specialists does reduce the stability of ecosystems (at least, once the
Ó 2016 The Authors
New Phytologist Ó 2016 New Phytologist Trust
Kress et al. (2009); Pei et al. (2011); Bezeng et al.
(2015); Clare (2014); Toju et al. (2014)
Shokralla et al. (2011)
community is reduced to complementary generalists) (Burkle et al.,
2013). With the goal of forecasting the effects of biodiversity loss,
metabarcoding techniques, combined with the collection of
georeferenced voucher specimens (Marques et al., 2013), provide
a means for extensive monitoring to reveal geographical hotspots of
declines in bee populations should pollinator communities be
assessed on a regular basis (Tang et al., 2015). Considering the
worrisome rate of range contraction of many bee species (Kerr et al.,
2015), these methods can be combined with extinction rates
estimated through phylogenetic methods to help address whether
the bee clades that are declining currently are the same as those that
have historically suffered declines (Hardy & Otto, 2014) as well as
what traits (such as levels of generalization in diet of pollen and
nectar sources) allow for resilience of bee species (Groom et al.,
2014).
To what degree is the extirpation of specialist species part of a
natural pattern of species turnover? We consider how DNA
barcodes can detect historical rates of loss of specialists in two
different ways: (1) using a historical ecological approach to detect
specialization from museum and herbarium records over decades to
centuries and (2) using a phylogenetic approach to detect
background extinction rates over evolutionary timescales. DNA
barcoding of pollen on collected specimens allows for, in addition
to the large-scale spatial analyses already described, large-scale
temporal analyses. By metabarcoding the pollen on herbarium (in
the case of floral stigmas) and museum (in the case of pollinators)
specimens, it will be possible to compare historical levels of
specialization to that of the present day and determine the number
New Phytologist (2016)
www.newphytologist.com
6 Forum
New
Phytologist
Viewpoints
and identity of species lost over time with increasing amounts of
anthropogenic disturbance (Scheper et al., 2014). This approach
appears feasible as researchers have been successful at detecting
DNA barcodes from museum specimens (Shokralla et al., 2011),
and DNA in pollen in particular appears amenable to deep time
synthesis (Suyama et al., 1996; Parducci et al., 2005).
Because commonly used plant barcode markers are phylogenetically informative, researchers have established that barcode
sequences are useful in plant community phylogenetics (Kress
et al., 2009; Pei et al., 2011; Bezeng et al., 2015). In such work, two
to three markers are typically sequenced, including a gene that is
broadly recognized for its utility for plant phylogenetics, rbcLa
(CBOL Plant Working Group, 2009; Kress et al., 2009; Pei et al.,
2011, 2015; Kuzmina et al., 2012; Bezeng et al., 2015). Thus, plant
phylogenies reconstructed from barcodes are generally found to be
accurate, i.e. when compared against phylogenies based upon more
evidence (Kress et al., 2009; Pei et al., 2011, 2015). By contrast,
single-marker (CO1) barcoding is generally performed for animals
due to its effectiveness for recognizing the majority of species as
revealed through traditional morphological or integrative methods
(e.g. Buide et al., 1998; Hebert et al., 2003a,b; Smith et al., 2006,
2008; Sheffield et al., 2009; Ratnasingham & Hebert, 2013;
Ondrejicka et al., 2014; Schmidt et al., 2015). Given the high rates
of molecular evolution, biased substitution patterns, and matrilineal mode of inheritance of the mitochondrial genome (Lin &
Danforth, 2004), barcode-based phylogenies for animals are
generally not expected to be as robust as nuclear-gene phylogenies
or the multi-gene phylogenies derived during plant barcoding.
However, animal barcode data can increase the taxon density in a
phylogenetic study when ‘backbone’ phylogenies constructed using
multi-gene data or phylogenomics approaches constrain the deeper
topology (e.g. genus or family-level relationships; see Trunz et al.,
2016), in combination with barcode data, which are available for
more species (Wilson, 2011; Boyle & Adamowicz, 2015). By
incorporating barcode markers, phylogenetic trees have better
phylogenetic resolution, which can yield more precise conclusions
regarding community structuring and processes (Kress et al., 2009;
Pei et al., 2011; Davies et al., 2012).
Application of phylogenetic comparative methods has indicated
that specialization is labile; evolutionary transitions from specialization to generalization are common (Abrahamczyk et al., 2014;
Vamosi et al., 2014b). However, extinction rate estimates from
these studies do suggest that specialized branches on the tree of life
are ephemeral relative to those of generalists (e.g. Colles et al.,
2009). Niche specialists are expected to have much smaller
geographical ranges (Williams et al., 2009) and be prone to greater
extinction rates due to lower global abundance (Packer et al.,
2005), yet there are few empirical tests in pollinator clades (Hardy
& Otto, 2014). Further study of plant and pollinator phylogenies
will allow us to estimate the degree of background extinction of
specialist lineages vs that of generalists as well as the propensity of
specialists to undergo transitions to generalization under certain
ecological conditions. With resolved phylogenetic trees we should
be able to examine whether lineages that have historically
experienced high levels of extinction are the same lineages that
are now experiencing population declines from anthropogenic
New Phytologist (2016)
www.newphytologist.com
disturbance. Relating these findings further to pollinator effectiveness in pollen delivery should result in a shortlist of efficient
pollinators that were historically robust but are now at risk, and
which might therefore increase the risk for the plants they visit.
Such approaches should form part of any decision-making,
concentrating conservation efforts where they are most needed.
Acknowledgements
The authors would like to thank the many pioneers of DNA
barcoding techniques. This work was supported by NSERC
Discovery Grants to J.C.V., S.J.A. and L.P. and an NSFC grant
(31670228) to Y-B.G.
Jana C. Vamosi1*, Yan-Bing Gong2, Sarah J. Adamowicz3 and
Laurence Packer4
1
Department of Biological Sciences, University of Calgary, 2500
University Drive NW, Calgary, AB T2N 1N4, Canada;
2
State Key Laboratory of Hybrid Rice, College of Life Sciences,
Wuhan University, Wuhan 430072, China;
3
Biodiversity Institute of Ontario & Department of Integrative
Biology, University of Guelph, 50 Stone Road East, Guelph,
ON N1G 2W1, Canada;
4
Department of Biology, York University, Toronto, ON M3J 1P3,
Canada
(*Author for correspondence: tel +1 403 210 9594;
email [email protected])
References
Abrahamczyk S, Souto-Vilaros D, Jorgensen PM, Renner SS. 2014. Escape from
extreme specialization: passionflowers, bats, and the Sword-billed hummingbird.
Proceedings of the Royal Society of London B: Biological Sciences 281: 20140888.
Adderley LA, Vamosi JC. 2015. Species and phylogenetic heterogeneity in
pollinator visitation affects selfing and seed production in an island system.
International Journal of Plant Sciences 176: 186–196.
Aguilar R, Ashworth L, Galetto L, Aizen MA. 2006. Plant reproductive
susceptibility to habitat fragmentation: review and synthesis through a metaanalysis. Ecology Letters 9: 968–980.
Aizenberg-Gershtein Y, Izhaki I, Halpern M. 2013. Do honeybees shape the
bacterial community composition in floral nectar. PLoS ONE 8: e67556.
Arceo-Gomez G, Ashman T-L. 2011. Heterospecific pollen deposition: does
diversity alter the consequences? New Phytologist 192: 738–746.
Ballantyne G, Baldock KCR, Willmer PG. 2015. Constructing more informative
plant–pollinator networks: visitation and pollen deposition networks in a
heathland plant community. Proceedings of the Royal Society of London B: Biological
Sciences 282: 20151130.
Bell KL, de Vere N, Keller A, Richardson RT, Gous A, Burgess KS, Brosi BJ. 2016.
Pollen DNA barcoding: current applications and future prospects. Genome 59:
629–640.
Bezeng SB, Davies TJ, Yessoufou K, Maurin O, Van der Bank M. 2015. Revisiting
Darwin’s naturalization conundrum: explaining invasion success of non-native
trees and shrubs in southern Africa. Journal of Ecology 103: 871–887.
Boyle EB, Adamowicz SJ. 2015. Community phylogenetics: assessing tree
reconstruction methods and the utility of DNA barcodes. PLoS ONE 10:
e0126662.
Bridle JR, Buckley J, Bodsorth EJ, Thomas CD. 2013. Evolution on the move:
specialization on widespread resources associated with rapid range expansion in
response to climate change. Proceedings of the Royal Society of London B: Biological
Sciences 281: 20131800.
Ó 2016 The Authors
New Phytologist Ó 2016 New Phytologist Trust
New
Phytologist
Brosi BJ, Briggs HM. 2013. Single pollinator species losses reduce floral fidelity and
plant reproductive function. Proceedings of the National Academy of Sciences, USA
110: 13044–13048.
Buide ML, Sanchez JM, Guitian J. 1998. Ecological characteristics of the flora of the
Northwest Iberian Peninsula. Plant Ecology 135: 1–8.
Burkle LA, Marlin JC, Knight TM. 2013. Plant–pollinator interactions over
120 years: loss of species, co-occurrence, and function. Science 339: 1611–
1615.
CBOL Plant Working Group. 2009. A DNA barcode for land plants. Proceedings of
the National Academy of Sciences, USA 106: 12794–12797.
Clare EL. 2014. Molecular detection of trophic interactions: emerging trends,
distinct advantages, significant considerations and conservation applications.
Evolutionary Applications 7: 1144–1157.
Clement WL, Donoghue MJ. 2012. Barcoding success as a function of phylogenetic
relatedness in Viburnum, a clade of woody angiosperms. BMC Evolutionary
Biology 12: 73.
Coissac E, Hollingsworth PM, Lavergne S, Taberlet P. 2016. From barcodes to
genomes: extending the concept of DNA barcoding. Molecular Ecology 25: 1423–
1428.
Colles A, Liow LH, Prinzing A. 2009. Are specialists at risk under environmental
change? Neoecological, paleoecological and phylogenetic approaches. Ecology
Letters 12: 849–863.
Davies TJ, Kraft NJB, Salamin N, Wolkovich EM. 2012. Incompletely resolved
phylogenetic trees inflate estimates of phylogenetic conservatism. Ecology 93:
242–247.
Davila YC, Elle E, Vamosi JC, Hermanutz L, Kerr JT, Lortie CJ, Westwood AR,
Woodcock TS, Worley A. 2012. Ecosystem services of pollinator diversity: a
review of the relationship with pollen limitation of plant reproduction. BotanyBotanique 90: 535–543.
Debarre F, Gandon S. 2010. Evolution of specialization in a spatially continuous
environment. Journal of Evolutionary Biology 23: 1090–1099.
Devictor V, Clavel J, Julliard R, Lavergne S, Mouillot D, Thuiller W, Venail P,
Villeger S, Mouquet N. 2010. Defining and measuring ecological specialization.
Journal of Applied Ecology 47: 15–25.
Fr€
und J, Dormann CF, Holzschuh A, Tscharntke T. 2013. Bee diversity effects on
pollination depend on functional complementarity and niche shifts. Ecology 94:
2042–2054.
Galimberti A, De Mattia F, Bruni I, Scaccabarozzi D, Sandionigi A, Barbuto M.
2014. A DNA barcoding approach to characterize pollen collected by honeybees.
PLoS ONE 9: e109363.
Gibbs J. 2010. Revision of the metallic species of Lasioglossum (Dialictus) in Canada
(Hymenoptera, Halictidae, Halictini). Zootaxa 3073: 1–382.
Gomez JM, Bosch J, Perfectti F, Fernandez J, Abdelaziz M. 2007. Pollinator
diversity affects plant reproduction and recruitment: the tradeoffs of
generalization. Oecologia 153: 597–605.
Groom SVP, Stevens MI, Schwarz MP. 2014. Parallel responses of bees to
Pleistocene climate change in three isolated archipelagos of the southwestern
Pacific. Proceedings of the Royal Society of London B: Biological Sciences 281:
20133293.
Harder LD, Barrett SCH. 1995. Mating cost of large floral displays in
hermaphrodite plants. Nature 373: 512–515.
Hardy N, Otto SP. 2014. Specialization and generalization in the
diversification of phytophagous insects: tests of the musical chairs and
oscillation hypotheses. Proceedings of the Royal Society of London B: Biological
Sciences 281: 1795.
Hebert PDN, Cywinska A, Ball SL, deWaard JR. 2003a. Biological identifications
through DNA barcodes. Proceedings of the Royal Society of London B: Biological
Sciences 270: 313–321.
Hebert PDN, Ratnasingham S, deWaard JR. 2003b. Barcoding animal life:
cytochrome c oxidae subunit 1 divergences among closely related species.
Proceedings of the Royal Society of London B: Biological Sciences 270: S96–S99.
Hollingsworth PM, Li D-Z, van der Bank M, Twyford AD. 2016. Telling plant
species apart with DNA: from barcodes to genomes. Philosophical Transactions of
the Royal Society B: Biological Sciences 371: 20150338.
Kerr JT, Pindar A, Galpern P, Packer L, Potts SG, Roberts SM, Rasmont P,
Schweiger O, Colla SR, Richardson LL et al. 2015. Climate change impacts on
bumblebees converge across continents. Science 349: 177–180.
Ó 2016 The Authors
New Phytologist Ó 2016 New Phytologist Trust
Viewpoints
Forum 7
King C, Ballantyne G, Willmer PG. 2013. Why flower visitation is a poor proxy for
pollination: measuring single-visit pollen deposition, with implications for
pollination networks and conservation. Methods in Ecology and Evolution 4: 811–
818.
Knight TM, Steets JA, Vamosi JC, Mazer SJ, Burd M, Campbell DR, Dudash MR,
Johnston MO, Mitchell RJ, Ashman T-L. 2005. Pollen limitation of plant
reproduction: pattern and process. Annual Review of Ecology, Evolution, and
Systematics 36: 467–497.
Kremen C, Williams NM, Thorp RW. 2002. Crop pollination from native bees at
risk from agricultural intensification. Proceedings of the National Academy of
Sciences, USA 99: 16812–16816.
Kress WJ, Erickson DL, Jones FA, Swenson NG, Perez R, Sanjur O, Bermingham
E. 2009. Plant DNA barcodes and a community phylogeny of a tropical forest
dynamics plot in Panama. Proceedings of the National Academy of Sciences, USA
106: 18621–18626.
Kress WJ, Wurdack KJ, Zimmer EA, Weigt LA, Janzen DH. 2005. Use of DNA
barcodes to identify flowering plants. Proceedings of the National Academy of
Sciences, USA 102: 8369–8374.
Kuzmina ML, Johnson KL, Barron HR, Hebert PDN. 2012. Identification of the
vascular plants of Churchill, Manitoba, using a DNA barcode library. BMC
Ecology 12: 25.
Li X, Yang Y, Henry RJ, Rossetto M, Wang Y, Chen S. 2015. Plant DNA
barcoding: from gene to genome. Biological Reviews 90: 157–166.
Lin C-P, Danforth B. 2004. How do insect nuclear and mitochondrial gene
substitution patterns differ? Insights from Bayesian analyses of combined datasets.
Molecular Phylogenetics and Evolution 30: 686–702.
Litsios G, Kostikova A, Salamin N. 2014. Host specialist clownfish are
environmental niche generalist. Proceedings of the Royal Society of London B:
Biological Sciences 281: 20133220.
Lloyd DG. 1982. Selection of combined versus separate sexes in seed plants.
American Naturalist 120: 571–585.
Mander L, Baker SJ, Belcher CM, Haselhorst DS, Rodriguez J, Thorn JL, Tiwari
S, Urrego DH, Wesseln CJ, Punyasena SW. 2014. Accuracy and consistency of
grass pollen identification by human analysts using electron micrographs of
surface ornamentation. Applications in Plant Sciences 2: 1400031.
Marcos JV, Nava R, Cristobal G, Redondo R, Escalante-Ramırez B, Bueno G,
Gonza lez-Porto A, Pardo C, Chung F et al. 2015. Automated pollen
Deniz O,
identification using microscopic imaging and texture analysis. Micron 68: 36–46.
Marques AC, Maronna MM, Collins AG. 2013. Putting GenBank data on the map.
Science 341: 1341–1341.
Memmott J, Waser NM, Price MV. 2004. Tolerance of pollination networks to
species extinctions. Proceedings of the Royal Society of London B: Biological Sciences
271: 2605–2611.
Muir JL, Vamosi JC. 2015. Invasive Scotch broom (Cytisus scoparius, Fabaceae) and
the pollination success of three Garry oak-associated plant species. Biological
Invasions 17: 2429–2446.
Ondrejicka DA, Locke SA, Morey K, Borisenko AV, Hanner RH. 2014. Status and
prospects of DNA barcoding in medically important parasites and vectors. Trends
in Parasitology 30: 582–591.
Packer L, Ali E, Dumesh S, Walker K. 2016. The identification of pollinators: where
are we and where should we go? In: Gemmill-Herren B, ed. Pollination services to
agriculture: sustaining and enhancing a key ecosystem service. London, UK: Taylor &
Francis, 57–73.
Packer L, Zayed A, Grixti JC, Ruz L, Owen RE, Vivallo F, Toro H. 2005.
Conservation genetics of potentially endangered mutualisms: reduced levels of
genetic variation in specialist versus generalist bees. Conservation Biology 19: 195–
202.
Parducci L, Suyama M, Lascoux M, Bennett KD. 2005. Ancient DNA from pollen:
a genetic record of population history in Scots pine. Molecular Ecology 14: 2873–
2882.
Pei N, Erickson D, Chen B, Ge X, Mi X, Swenson N, Zhang J-L, Jones F, Huang CL, Ye W et al. 2015. Closely-related taxa influence woody species discrimination
via DNA barcoding: evidence from global forest dynamics plots. Scientific Reports
5: 15127.
Pei N, Lian J-Y, Erickson D, Swenson N, Kress W, Ye W-H, Ge X-J. 2011.
Exploring tree-habitat associations in a Chinese subtropical forest plot using a
molecular phylogeny generated from DNA barcode loci. PLoS ONE 6: e21273.
New Phytologist (2016)
www.newphytologist.com
8 Forum
Viewpoints
Popic TJ, Davila YC, Wardle GM. 2013. Evaluation of common methods for
sampling invertebrate pollinator assemblages: net sampling out-perform pan
traps. PLoS ONE 8: e66665.
Pornon A, Escaravage N, Burrus M, Holota H, Khimoun A, Mariette J, Pellizzari
C, Iribar A, Etienne R, Taberlet PDN et al. 2016. Using metabarcoding to reveal
and quantify plant–pollinator interactions. Scientific Reports 6: 27282.
Ratnasingham S, Hebert PDN. 2007. BOLD: the barcode of life data system (www.
barcodinglife.org). Molecular Ecology Notes 7: 355–364.
Ratnasingham S, Hebert P. 2013. A DNA-based registry for all animal species: the
barcode index number (BIN) system. PLoS ONE 8: e66213.
Richardson RT, Lin C-H, Quijia J, Riusech N, Goodell K, Johnson R. 2015a.
Rank-based characterization of pollen assemblages collected by honey bees using a
multi-locus metabarcoding approach. Applications in Plant Sciences 3: 1500043.
Richardson RT, Lin C-H, Sponsler DB, Quijia JO, Goodell K, Johnson RM.
2015b. Application of ITS2 metabarcoding to determine the provenance of
pollen collected by honey bees in an agroecosystem. Applications in Plant Sciences
3: 1400066.
Roslin T, Majaneva S. 2016. The use of DNA barcodes in food web construction –
terrestrial and aquatic ecologists unite! Genome 59: 603–628.
Scheper J, Reemer M, van Kats R, Ozinga WA, van der Linden GTJ, Schaminee
JHJ, Siepel H, Kleijn D. 2014. Museum specimens reveal loss of pollen host
plants as key factor driving wild bee decline in the Netherlands. Proceedings of the
National Academy of Sciences, USA 111: 17552–17557.
Schmidt S, Schmid-Egger C, Moriniere J, Haszprunar G, Hebert PDN. 2015.
DNA barcoding largely supports 250 years of classical taxonomy: identifications
for Central European bees (Hymenoptera, Apoidea partim). Molecular Ecology
Resources 15: 985–1000.
Schnell IB, Fraser M, Willerslev E, Gilbert MTP. 2010. Characterisation of insect
and plant origins using DNA extracted from small volumes of bee honey.
Arthropod-Plant Interactions 4: 107–116.
Schuett EM, Vamosi JC. 2010. Phylogenetic community context influences pollen
delivery to Allium cernuum. Evolutionary Biology 37: 19–28.
Sheffield CS, Hebert PDN, Kevan P, Packer L. 2009. DNA barcoding a regional
bee (Hymenoptera: Apoidea) fauna and its potential for ecological studies.
Molecular Ecology Resources 9(Suppl. 1): 196–207.
Shokralla S, Zhou X, Janzen DH, Hallwachs W, Landry J-F, Jacobus LM,
Hajibabaei M. 2011. Pyrosequencing for mini-barcoding of fresh and old
museum specimens. PLoS ONE 6: e21252.
Sickel W, Ankenbrand MJ, Grimmer G, Holzschuh A, H€a rtel S, Lanzen J, SteffanDewenter I, Keller A. 2015. Increased efficiency in identifying mixed pollen
samples by meta-barcoding with a dual-indexing approach. BMC Ecology 15: 1–9.
Sjodin NE. 2007. Pollinator behavioural responses to grazing intensity. Biodiversity
and Conservation 16: 2103–2121.
Smith MA, Eveleigh ES, McCann KS, Merilo MT, McCarthy PC, Van Rooyen KI.
2011. Barcoding a quantified food web: crypsis, concepts, ecology and
hypotheses. PLoS One 6: e14424.
Smith MA, Rodriguez JJ, Whitfield JB, Deans AR, Janzen DH, Hallwachs W,
Hebert PDN. 2008. Extreme diversity of tropical parasitoid wasps exposed by
iterative integration of natural history, DNA barcoding, morphology, and
collections. Proceedings of the National Academy of Sciences, USA 105: 12359–
12364.
Smith MA, Woodley NE, Janzen DH, Hallwachs W, Hebert PDN. 2006. DNA
barcodes reveal cryptic host-specificity within the presumed polyphagous
members of a genus of parasitoid flies (Diptera: Tachinidae). Proceedings of the
National Academy of Sciences, USA 103: 3657–3662.
Stork NE, McBrooma J, Gelyb C, Hamilton AJ. 2015. New approaches narrow
global species estimates for beetles, insects, and terrestrial arthropods. Proceedings
of the National Academy of Sciences, USA 112: 7519–7523.
New Phytologist (2016)
www.newphytologist.com
New
Phytologist
Suyama Y, Kawamuro K, Kinoshita I, Yoshimura K, Tsumura Y, Takahara H.
1996. DNA sequence from a fossil pollen of Abies spp. from Pleistocene peat.
Genes & Genetic Systems 71: 145–149.
Tang M, Hardman CJ, Yinqiu J, Meng G, Liu L, Tan T, Yang S, Moss ED,
Wang J, Yang C et al. 2015. High-throughput monitoring of wild bee
diversity and abundance via mitogenomics. Methods in Ecology and Evolution
6: 1034–1043.
Toju H, Guimar~aes PR, Olesen JM, Thompson JN. 2014. Assembly of complex
plant–fungus networks. Nature Communications 5: 5273.
Trunz V, Packer L, Vieu J, Arrigo N, Praz CJ. 2016. Comprehensive phylogeny,
biogeography and new classification of the diverse bee tribe Megachilini: can we
use DNA barcodes in phylogenies of large genera? Molecular Phylogenetics and
Evolution 103: 245–259.
Tur C, Castro-Urgal R, Traveset A. 2013. Linking plant specialization to
dependence in interactions for seed set in pollination networks. PLoS ONE 8:
e78294.
Ushio M, Yamasaki E, Takasu H, Nagono AJ, Fujinaga S, Honjo MN, Ikemoto M,
Sakai S, Kudoh H. 2015. Microbial communities on flower surfaces act as
signatures of pollinator visitation. Scientific Reports 5: 8695.
Vamosi JC, Moray CM, Garcha NK, Chamberlain SA, Mooers AØ. 2014a.
Pollinators visit related plant species across 29 plant–pollinator networks. Ecology
and Evolution 4: 2303–2315.
Vamosi JC, Renner S, Armbruster WS. 2014b. Evolutionary ecology of
specialisation: insights from phylogenetic analysis. Proceedings of the Royal Society,
Series B 281: 20142004.
Vamosi JC, Sargent RD, Elle E. 2012. Pollination biology research in Canada:
perspectives on a mutualism at different scales. Botany-Botanique 9: v–vi.
Vamosi JC, Steets JA, Ashman T-L. 2013. Drivers of pollen limitation:
macroecological interactions between breeding system, rarity, and diversity. Plant
Ecology and Diversity 6: 171–180.
Weiner CN, Werner M, Linsenmair KE, Bl€
uthgen N. 2014. Land-use impacts on
plant–pollinator networks: interaction strength and specialization predict
pollinator declines. Ecology 95: 466–474.
Whitlock MC. 1996. The Red Queen beats the Jack-of-all-trades: the limitations on
the evolution of phenotypic plasticity and niche breadth. American Naturalist
148: S65–S77.
Widmer A, Cozzolino S, Pellegrino G, Soliva M, Dafni A. 2000. Molecular analysis
of orchid pollinaria and pollinaria-remains found on insects. Molecular Ecology 9:
1911–1914.
Williams SE, Williams YM, VanDerWal J, Isaac JL, Shoo LP, Johnson CN. 2009.
Ecological specialization and population size in a biodiversity hotspot: how rare
species avoid extinction. Proceedings of the National Academy of Sciences, USA 17:
19737–19741.
Wilson JJ. 2011. Assessing the value of DNA barcodes for molecular phylogenetics:
effect of increased taxon sampling in Lepidoptera. PLoS ONE 6: e24769.
Wilson EE, Sidhu CS, LeVan KE, Holway DA. 2010. Pollen foraging behaviour of
solitary Hawaiian bees revealed through molecular pollen analysis. Molecular
Ecology 19: 4823–4829.
Zayed A, Packer L, Grixti JC, Ruz L, Owen RE, Toro H. 2005. Increased genetic
differentiation in a specialist versus a generalist bee: implications for conservation.
Conservation Genetics 5: 1017–1026.
Key words: diversity, ecosystem function, extinction, metabarcoding, specialization,
speciation, stability.
Ó 2016 The Authors
New Phytologist Ó 2016 New Phytologist Trust